LoopX

LoopX is an open, local-first control plane that maintains durable state and goals for long-horizon AI agents.

LoopX is a control plane with a durable state kernel designed for long-horizon agents and personal agent teams. It connects projects, plans work as todos, and sets up heartbeats to keep tasks moving across sessions. The software works with runtimes like Codex and Claude Code, providing a local-first workspace for goals, tasks, files, and schedules. LoopX is built for developers and researchers running multi-day engineering, research, or operations workflows. It sets itself apart by maintaining state authoritatively in a compact layer rather than acting as another agent framework or provider-specific orchestration runtime.

LoopX at a glance

Best for: Software engineer
Runs on: Web, macOS, Windows, CLI, API, self-hosted, open source
Pricing: Free
Released: May 2026
Category: AI agent tool
Website: github.com/loopx-project/loopx

Key features

Govern durable control state across sessions
Maintain long-horizon goals and evidence
Support Codex and Claude Code runtimes
Manage tasks through an agent-native Kanban board
Handle peer ownership and recoverable handoffs
Configure goal capabilities and preview changes
Queue messages and steer live agent turns

Use it for

Keep coding or research agents working across sessions
Manage personal projects, schedules, and decisions
Let agents collaborate and deliver verifiable results
Run multi-day engineering and experiment objectives

Alternatives to LoopX

Loopy — Provides repeatable prompts with checks, proof, and stopping conditions.
LongHorizon Harness — Maintains durable task state and executes reliable, verifiable workflows over many hours.
CodePilot — Consolidates multiple AI agents and local model connections in one workspace.
Loop Engineering — Design agent loops to discover work and verify results automatically.
Coze Loop — It offers open-source full-lifecycle management for AI agents.
Agentloop — Provides a clean agent loop with provider-agnostic LLM access, hooks, and safety budgets.
AgentOps — Track token usage, costs, and traces across 400 plus LLMs and agent frameworks.
Lever — Agents iterate autonomously against set metrics and win criteria for hours.
AGiXT — Builds autonomous agents with integrated memory and web capabilities.
Mission Control — Consolidates multi-agent operations, task boards, and telemetry into one self-hosted dashboard.
Kungfu — Maintains work state and context when switching between different AI agents.
Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps

All LoopX alternatives